scholarly journals Platelet-Derived Proteases ADAM10 and ADAM17 Impair NK Cell Immunosurveillance of Metastasizing Tumor Cells By Diminishing NKG2D Ligand Surface Expression

Blood ◽  
2014 ◽  
Vol 124 (21) ◽  
pp. 4164-4164
Author(s):  
Stefanie Raab ◽  
Korbinian Nepomuk Kropp ◽  
Alexander Steinle ◽  
Gerd Klein ◽  
Lothar Kanz ◽  
...  

Abstract Introduction: NK cells are cytotoxic lymphocytes the reactivity of which is governed by the principles of ‘missing-self’ and “induced-self’ recognition. This implies that cells with low or absent expression of MHC class I and/or stress-induced expression of ligands for activating receptors like NKG2D (NKG2DL) are preferentially eliminated by NK cells. We and others reported previously that NKG2DL undergo proteolytic cleavage (i.e. shedding) which promotes evasion from NKG2D-mediated tumor immunosurveillance. Notably, the reactivity of NK cells is also influenced by various other components of the hematopoietic system, and we and others provided evidence for the involvement of platelets in tumor immune evasion including impairment of NKG2D-mediated immune surveillance (e.g., Kopp et al., Cancer Res 2009, Raab et al., Blood 2013 122:3488). Here we extend our recent findings and provide further data how platelets affect immunostimulatory NKG2D-NKG2DL interaction. Methods: Tumor cells were incubated with platelets from healthy donors or treated with platelet-derived soluble factors (releasate) obtained by stimulation of platelets with known platelet agonists including ADP and thrombin. NKG2DL and ADAM10 as well as ADAM17 surface expression on tumor cells and platelets, respectively, was measured by FACS, while release of NKG2DL was determined by ELISA. ADAM10 and 17 protein levels were assessed by immunoblotting. NK cell lysis of tumor cells in the presence or absence of coating platelets or platelet releasate was determined by chromium release assays. Results: We found that interaction of platelets with tumor cells resulted in substantially reduced NKG2DL expression on the surface of the malignant cells, which was paralleled by enhanced release of soluble NKG2DL. Similar albeit weaker effects were observed upon treatment of tumor cells with platelet releasate, indicating that platelet-derived factors mediate NKG2DL shedding from the tumor cell surface. Notably, ADAM10 and ADAM17, the known sheddases of NKG2DL, were found to be expressed on the platelet surface, and sheddases could also be detected in platelet releasate, pointing to an involvement in platelet-mediated NKG2DL shedding. Diminished NKG2DL surface expression resulted in reduced NKG2D-mediated NK cell cytotoxicity as revealed by blocking experiments using NKG2D antibody and F(ab)2 fragments specific for the modulated NKG2DL. Conclusion: We propose that induction of NKG2DL shedding constitutes a novel mechanism by which the interaction of platelets with metastasizing tumor cells impairs NK cell immunosurveillance. Disclosures No relevant conflicts of interest to declare.

Blood ◽  
2010 ◽  
Vol 116 (21) ◽  
pp. 3193-3193 ◽  
Author(s):  
Theresa Placke ◽  
Lothar Kanz ◽  
Helmut R. Salih ◽  
Hans-Georg Kopp

Abstract Abstract 3193 NK cells as part of the innate immune system substantially contribute to cancer immune surveillance. They prevent tumor progression and metastasis due to their ability to mediate cellular cytotoxicity and to produce cytokines like IFN-γ, which, among others, stimulates subsequent adaptive immune responses. NK reactivity results from an integrative response emerging upon recognition of multiple ligands for activating and inhibitory NK cell receptors including various members of the TNFR family. Apart from the direct interaction with their target cells, NK cell activity is further influenced by the reciprocal interplay with various other hematopoietic cells like e.g. dendritic cells. Metastatic tumor spread in experimental animals is dramatically reduced in thrombopenic mice. Additional depletion of NK cells reverses this effect, indicating that platelets may impair NK anti-tumor reactivity. However, the underlying mechanisms have not been fully elucidated, especially in humans. Recently, we demonstrated that NK anti-tumor immunity is impaired by platelet-derived TGF-β, which is released upon interaction of platelets with tumor cells (Kopp et al., Cancer Res. 2009). Here we report that the ligand for the TNFR family member GITR (GITRL) is upregulated on megakaryocytes during maturation resulting in substantial GITRL expression by platelets. Since we recently identified GITR as inhibitory NK receptor involved in tumor immune escape (e.g., Baltz et al., Blood 2008, Baessler et al., Cancer Res. 2009) we investigated how platelet-derived GITRL influences platelet function and NK immune surveillance. Signaling via GITRL into platelets upon interaction with NK-expressed GITR or recombinant GITR-Ig fusion protein did not alter platelet activation as revealed by analysis of the activation marker CD62P and release of TGF-β. Interestingly, we found that GITRL-negative tumor cells rapidly get coated by platelets, which confers a seemingly GITRL-positive phenotype. “GITRL pseudoexpression” on tumor cells caused a substantial reduction of NK cell cytotoxicity and cytokine production. This reduced NK reactivity was not due to induction of apoptosis via GITR and could be restored by addition of a blocking GITR antibody. Thus, coating of tumor cells by platelets inhibits NK reactivity, which is in part mediated by platelet-derived GITRL. Our data provide a functional basis for the previously observed finding that platelets increase metastasis i.e. by enabling evasion of tumor cells from NK-mediated immune surveillance. Disclosures: No relevant conflicts of interest to declare.


Blood ◽  
2011 ◽  
Vol 118 (21) ◽  
pp. 245-245
Author(s):  
Benjamin J Schmiedel ◽  
Tina Baessler ◽  
Miyuki Azuma ◽  
Lothar Kanz ◽  
Helmut R. Salih

Abstract Abstract 245 The TNF family member RANKL and its receptors RANK and osteoprotegerin (OPG) are key regulators of bone remodelling, but have also been shown to influence progression of malignancies like breast cancer (Tan et al., Nature 2011), myeloma (Sordillo et al., Cancer 2003) and CLL (Secchiero et al. J Cell Physiol. 2006). NK cells are cytotoxic lymphocytes that play an important role in tumor immune surveillance especially of hematopoietic malignancies. Their reactivity is influenced by a variety of activating and inhibitory molecules expressed by their target cells including several members of the TNF family. Recently, we reported that RANK, upon interaction with RANKL which can be expressed by malignant hematopoietic cells, mediates signals that impair NK reactivity (Schmiedel et al., Blood 2010 116,21:893–893). Here we extended these analyses and comprehensively studied the expression and immunomodulatory function of RANKL in leukemia. Analysis of primary leukemia cells revealed substantial RANKL surface expression in a high proportion of the investigated cases (AML, 47 of 65 (72%); ALL, 16 of 21 (76%); CML, 6 of 10 (60%); CLL, all 54 (100%)). Signaling via surface-expressed RANKL into the malignant cells mediated the release of cytokines like TNF, IL-6, IL-8 and IL-10 which have been shown to act as autocrine and paracrine growth and survival factors in leukemia. Moreover, the factors released upon RANKL signaling upregulated RANK expression on NK cells. In line, NK cells from leukemia patients (n=75) displayed significantly (p<0.001, Mann-Whitney U-test) higher RANK expression compared to healthy controls (n=30) confirming our notion that RANK-RANKL interaction may contribute to leukemia pathophysiology. We further found that RANK-RANKL interaction, beyond directly inhibiting NK cell function via RANK, may contribute to evasion of leukemia cells from NK immunosurveillance by creating an NK inhibitory cytokine milieu. This was revealed by impaired cytotoxicity and degranulation in response to leukemia targets following exposure of the NK cells to the factors released upon RANKL signaling by leukemia cells. Notably, the RANKL-mediated cytokine release of leukemia cells could be disrupted by the clinically approved RANKL antibody Denosumab/AMG162. Thus, RANKL signaling may trigger a “vicious cycle” comprising of release of immunosuppressive cytokines and also upregulation of RANK on NK cells. The latter both directly inhibits NK reactivity and may result in augmented RANKL signaling into leukemia cells. Our data suggest that therapeutic modulation of the RANK/RANKL system e.g. with Denosumab/AMG162, which is approved for treatment of osteolysis, may be a promising strategy to reinforce NK reactivity against hematopoietic malignancies. Disclosures: No relevant conflicts of interest to declare.


Blood ◽  
2014 ◽  
Vol 124 (21) ◽  
pp. 5810-5810
Author(s):  
Elisabetta Todisco ◽  
Ilaria Turin ◽  
Federica Ferulli ◽  
Matteo Tanzi ◽  
Silvia Brugnatelli ◽  
...  

Abstract Background The failure of conventional treatment and target therapies to significantly improve outcomes in metastatic colorectal cancer (mCRC) has prompted the development of immune-based therapies, including natural killer (NK) cells-based strategies. NK cells can kill target cells directly, as well as mediate antibody (Ab)-dependent cellular cytotoxicity (ADCC) via the membrane receptor FcgRIII, which binds to the Fc portion of IgG Ab. This feature makes them of particular importance in the immunotherapy of mCRC patients harboring the KRAS mutation that cannot benefit from administration of anti-epidermal growth factor receptor (EGFR) drugs. In this study we evaluated the capacity of patients derived NK cells, either resting or after cytokine activation, to lyse autologous mCRC cells. mCRC cells were analyzed for expression of ligands for adhesion and triggering NK receptors involved in their recognition and killing. We also evaluated whether KRAS mutated mCRC cells were susceptible to anti-EGFR-induced ADCC mediated by NK cells. Patients and methods. After obtaining informed signed consent, 25 mCRC patients have been enrolled to date. Tumor cells were disaggregated by GentleMACS Dissociator (Miltenyi Biotec,Germany), in vitro expanded and analyzed for the expression of ligands for NK triggering receptors. Ligands expression was evaluated by cytofluorimetric analysis and gene expression by qPCR on mCRC cultured cells and by immunohistochemistry in sections of samples embedded in paraffin. Resting and IL-2- or IL-15-activated NK cells, were analyzed for expression of triggering and inhibitory receptors and for their ability to kill autologous mCRC cells alone or after incubation with anti-EGFR monoclonal antibodies (mAbs) in a 51Cr release cytotoxicity assay. Results. Tumor cells were successfully expanded from 21 of 25 samples. Experiments performed in 10 patients showed the inability of patients resting NK cells to lyse mCRC cells (< 10% at effector:target ratio (E:T) of 20:1. Cytokine overnight (ON) activation resulted in an increased NK cytotoxic activity (IL-2: mean 28%; range:10-71; and IL-15: mean 40%; range 16-76 at E:T ratio of 20:1). Additional days of NK cell activation were able to further enhance their lytic capability. In resting NK cells the mean surface expression of activating receptors DNAM-1 and NKG2D was 79%, (range 75-91) and 39 % (range 29-58), respectively. The latter was up-regulated by ON cytokine activation (IL-2: mean 53%, range 48-70; IL-15: mean 70%, range 63-87). Among activating natural cytotoxicity receptors (NCRs), NKp46 is highly expressed both on resting and activated cells (>90%), while low surface expression of NKp30 and NKp44 was documented on resting NK cells (<10%). Their expression could be slightly up-regulated in particular after IL-15 ON activation (mean: 13%, range 9-30). Molecular analysis demonstrated a sizeable gene expression of NK ligands PVR, Nectin-2 and MICA/B on mCRC cells. Cytofluorimetric analysis showed that PVR and Nectin-2 are expressed in a vast majority of cultured mCRC cells (mean 65%, range 60-76; and mean 78%, range 70-98), respectively, while MICA/B were less expressed (mean 18%, range 15-25). The incubation of mCRC cells with anti-EGFR mAbs increased their susceptibility to NK-mediated lysis irrespective of KRAS status of tumor cells. The average rate of increase was greater using resting NK cells as effectors. The ongoing experiments comparing mCRC ligands expression evaluated on cultured tumor cells versus paraffin embedded sections will clarify whether results obtained in vitrocould be translated to the clinical setting. Conclusions. The evidence that ex vivo activated autologous NK cells are able to lyse patients tumor cells can offer new therapeutic options to a cohort of mCRC patients with a poor prognosis. Disclosures No relevant conflicts of interest to declare.


Pharmaceutics ◽  
2021 ◽  
Vol 13 (4) ◽  
pp. 525
Author(s):  
Kwang-Soo Kim ◽  
Dong-Hwan Kim ◽  
Dong-Hyun Kim

Among various immunotherapies, natural killer (NK) cell cancer immunotherapy using adoptive transfer of NK cells takes a unique position by targeting tumor cells that evade the host immune surveillance. As the first-line innate effector cell, it has been revealed that NK cells have distinct mechanisms to both eliminate cancer cells directly and amplify the anticancer immune system. Over the last 40 years, NK cell cancer immunotherapy has shown encouraging reports in pre-clinic and clinic settings. In total, 288 clinical trials are investigating various NK cell immunotherapies to treat hematologic and solid malignancies in 2021. However, the clinical outcomes are unsatisfying, with remained challenges. The major limitation is attributed to the immune-suppressive tumor microenvironment (TME), low activity of NK cells, inadequate homing of NK cells, and limited contact frequency of NK cells with tumor cells. Innovative strategies to promote the cytolytic activity, durable persistence, activation, and tumor-infiltration of NK cells are required to advance NK cell cancer immunotherapy. As maturing nanotechnology and nanomedicine for clinical applications, there is a greater opportunity to augment NK cell therapeutic efficacy for the treatment of cancers. Active molecules/cytokine delivery, imaging, and physicochemical properties of nanoparticles are well equipped to overcome the challenges of NK cell cancer immunotherapy. Here, we discuss recent clinical trials of NK cell cancer immunotherapy, NK cell cancer immunotherapy challenges, and advances of nanoparticle-mediated NK cell therapeutic efficacy augmentation.


2020 ◽  
Vol 11 ◽  
Author(s):  
Stefanie Maurer ◽  
Lucas Ferrari de Andrade

Natural killer (NK) cells recognize and kill tumor cells via germ-line encoded receptors and polarized degranulation of cytotoxic molecules, respectively. As such, NK cells help to inhibit the development of cancers. The activating receptor NKG2D induces NK cell-mediated killing of metastasizing tumor cells by recognition of the stress-induced ligands MICA, MICB, and ULBP1-6. However, platelets enable escape from this immune surveillance mechanism by obstructing the interactions between NK cells and tumor cells or by cleaving the stress-induced ligands. It is also being increasingly appreciated that NK cells play additional roles in cancer immunity, including chemokine-mediated recruitment of antigen presenting cells in the tumor microenvironment that is followed by generation of adaptive immunity. However, the NK cell interplays with dendritic cells, and macrophages are extremely complex and involve molecular interactions via NKG2D and cytokine receptors. Specifically, NKG2D-mediated chronic interaction between NK cells and tumor-infiltrating macrophages causes immune suppression by differentiating NK cells toward a dysfunctional state. Here we discuss the underlying mechanisms of NK cell control by platelets and myeloid cells with focus on NKG2D and its ligands, and provide a timely perspective on how to harness these pathways with novel immunotherapeutic approaches.


Blood ◽  
2010 ◽  
Vol 116 (21) ◽  
pp. 2164-2164
Author(s):  
Benjamin J Schmiedel ◽  
Constantin M Wende ◽  
Tina Baessler ◽  
Carolin Scheible ◽  
Stefan Wirths ◽  
...  

Abstract Abstract 2164 NK cells play an important role in tumor immunosurveillance, especially of leukemia. Their reactivity is governed by various activating and inhibitory molecules expressed by their targets including multiple members of the TNF family. The TNF family member Receptor Activator of NF-κB ligand (RANKL) and its receptors RANK and osteoprotegerin (OPG) are key regulators of bone remodelling, but recently have also been shown to influence progression of hematopoetic malignancies. Here we studied the yet unkown role of the RANK/RANKL molecule system in NK cells and their reactivity against acute myeloid leukemia (AML). Primary leukemia cells from AML patients were found to substantially express RANKL mRNA and surface protein in 75% of the investigated cases (n=40). Reverse signaling via surface-expressed RANKL into AML blasts induced the release of soluble factors including the immunoregulatory cytokines TNF and IL-10, which impaired NK cell anti-tumor reactivity. Moreover, we observed upregulation of RANK on NK cells among PBMC of healthy donors upon exposure to IL-10. This was not caused by direct effects on NK cells, but was rather due to yet unidentified factors released by monocytes among the PBMC upon IL-10 exposure and could be prevented by the activating cytokine IL-2. Furthermore, functional experiments with NK cells and RANKL transfectants or RANKL-negative controls revealed that forward signaling into RANK-expressing NK cells by tumor-expressed RANKL also directly impaired NK cytotoxicity and IFN-γ production. In line, blocking RANK-RANKL interaction using anti-RANKL antibodies or RANK-Fc fusion protein increased cytotoxicity and cytokine production of allogenic NK cells in cultures with RANKL-positive primary AML cells. Our data indicate that RANKL expression enables immune evasion of leukemia cells both by directly inhibiting reactivity of RANK-expressing NK cells and by orchestrating a reciprocal interplay between AML cells, monocytes and NK cells resulting in an immunosuppressive cytokine milieu. Thus, therapeutic modulation of the RANK/RANKL system, e.g. with Denosumab/AMG162, which is presently being evaluated for treatment of both non-malignant and malignant osteolysis, holds promise to reinforce NK reactivity against hematopoietic malignancies. Disclosures: No relevant conflicts of interest to declare.


Blood ◽  
2010 ◽  
Vol 116 (21) ◽  
pp. 583-583 ◽  
Author(s):  
Theresa Placke ◽  
Hans-Georg Kopp ◽  
Martin Schaller ◽  
Gundram Jung ◽  
Lothar Kanz ◽  
...  

Abstract Abstract 583 NK cells are a central component of the cytotoxic lymphocyte compartment capable of lysing tumor cells without prior immune sensitization of the host. The mechanisms leading to activation of NK reactivity are described by the principles of ‘missing-self' and ‘induced-self', which imply that cells with a low or absent expression of MHC class I (‘missing-self') and/or a stress-induced expression of ligands of activating NK receptors like e.g. NKG2D (‘induced-self') are preferentially recognized and eliminated by NK cells. Thus, a balance of various activating and inhibitory signals determines whether NK cell responses are initiated or not. Tumor cells often downregulate expression of MHC class I to evade T cell-mediated immune surveillance, which results in enhanced NK susceptibility. Besides the direct interaction with their target cells, NK activity is further influenced by the reciprocal interplay with various other hematopoietic cells. We and others demonstrated previously that thrombocytopenia inhibits metastasis in murine models, which is reversed by additional depletion of NK cells (e.g., Jin et al., Nature Med. 2006, Palumbo et al., Blood 2005). However, the mechanisms by which platelets impair NK-tumor interaction are largely unclear, especially in humans. Recently we reported that platelets release TGF-β upon interaction with tumor cells causing downregulation of NKG2D on NK cells, which impairs anti-tumor immunity by disturbing the principle of “induced self” (Kopp et al., Cancer Res. 2009). Here we demonstrate that platelets further enable tumor cells to evade NK cell immune surveillance by preventing detection of “missing self”: We found that tumor cells rapidly get coated in the presence of platelets, the latter expressing large amounts of MHC class I on their surface. In case of MHC class I-negative or -low cancer cells, this process results in MHC class I “pseudoexpression” on the tumor cell surface as revealed by flow cytometry, immunofluorescent staining, and electron microscopy. Platelet-derived MHC class I was found to inhibit the reactivity of autologous NK, both upon activation with cytokines and, most importantly, in cultures with platelet-coated tumor cells. Using constitutively MHC class I-negative/low tumor cells we found that blocking MHC class I restored NK cytotoxicity and IFN-γ production against platelet-coated tumor cells, but did not alter NK reactivity against the tumor cells in the absence of coating platelets. Taken together, our data indicate that platelets enable a molecular mimicry of tumor cells, allowing the latter to downregulate MHC class I in order to escape T cell immunity without inducing sufficient NK tumor immune surveillance due to conferred platelet-mediated “pseudo self”. Disclosures: No relevant conflicts of interest to declare.


Cancers ◽  
2021 ◽  
Vol 13 (2) ◽  
pp. 239
Author(s):  
Kawaljit Kaur ◽  
Tahmineh Safaie ◽  
Meng-Wei Ko ◽  
Yuhao Wang ◽  
Anahid Jewett

Tumor cells are known to upregulate major histocompatibility complex-class I chain related proteins A and B (MICA/B) expression under stress conditions or due to radiation exposure. However, it is not clear whether there are specific stages of cellular maturation in which these ligands are upregulated or whether the natural killer (NK) cells differentially target these tumors in direct cytotoxicity or antibody-dependent cell cytotoxicity (ADCC). We used freshly isolated primary and osteoclast (OCs)-expanded NK cells to determine the degree of direct cytotoxicity or of ADCC using anti-MICA/B monoclonal antibodies (mAbs) against oral stem-like/poorly-differentiated oral squamous cancer stem cells (OSCSCs) and Mia PaCa-2 (MP2) pancreatic tumors as well as their well-differentiated counterparts: namely, oral squamous carcinoma cells (OSCCs) and pancreatic PL12 tumors. By using phenotypic and functional analysis, we demonstrated that OSCSCs and MP2 tumors were primary targets of direct cytotoxicity by freshly isolated NK cells and not by ADCC mediated by anti-MICA/B mAbs, which was likely due to the lower surface expression of MICA/B. However, the inverse was seen when their MICA/B-expressing differentiated counterparts, OSCCs and PL12 tumors, were used in direct cytotoxicity and ADCC, in which there was lower direct cytotoxicity but higher ADCC mediated by the NK cells. Differentiation of the OSCSCs and MP2 tumors by NK cell-supernatants abolished the direct killing of these tumors by the NK cells while enhancing NK cell-mediated ADCC due to the increased expression of MICA/B on the surface of these tumors. We further report that both direct killing and ADCC against MICA/B expressing tumors were significantly diminished by cancer patients’ NK cells. Surprisingly, OC-expanded NK cells, unlike primary interleukin-2 (IL-2) activated NK cells, were found to kill OSCCs and PL12 tumors, and under these conditions, we did not observe significant ADCC using anti-MICA/B mAbs, even though the tumors expressed a higher surface expression of MICA/B. In addition, differentiated tumor cells also expressed higher levels of surface epidermal growth factor receptor (EGFR) and programmed death-ligand 1(PDL1) and were more susceptible to NK cell-mediated ADCC in the presence of anti-EGFR and anti-PDL1 mAbs compared to their stem-like/poorly differentiated counterparts. Overall, these results suggested the possibility of CD16 receptors mediating both direct cytotoxicity and ADCC, resulting in the competitive use of these receptors in either direct killing or ADCC, depending on the differentiation status of tumor cells and the stage of maturation and activation of NK cells.


2021 ◽  
Vol 12 ◽  
Author(s):  
Lukman O. Afolabi ◽  
Jiacheng Bi ◽  
Xuguang Li ◽  
Adeleye O. Adeshakin ◽  
Funmilayo O. Adeshakin ◽  
...  

Histone deacetylases (HDAC) are frequently overexpressed in tumors, and their inhibition has shown promising anti-tumor effects. However, the synergistic effects of HDAC inhibition with immune cell therapy have not been fully explored. Natural killer (NK) cells are cytotoxic lymphocytes for anti-tumor immune surveillance, with immunotherapy potential. We showed that a pan-HDAC inhibitor, panobinostat, alone demonstrated anti-tumor and anti-proliferative activities on all tested tumors in vitro. Additionally, panobinostat co-treatment or pretreatment synergized with NK cells to mediate tumor cell cytolysis. Mechanistically, panobinostat treatment increased the expression of cell adhesion and tight junction-related genes, promoted conjugation formation between NK and tumor cells, and modulates NK cell-activating receptors and ligands on tumor cells, contributing to the increased tumor cytolysis. Finally, panobinostat therapy led to better tumor control and synergized with anti-PD-L1 therapy. Our data highlights the anti-tumor potential of HDAC inhibition through tumor-intrinsic toxicity and enhancement of NK –based immunotherapy.


Blood ◽  
2007 ◽  
Vol 110 (11) ◽  
pp. 2744-2744 ◽  
Author(s):  
Maria Berg ◽  
Andreas Lundqvist ◽  
Yong Fan ◽  
J. Philip McCoy ◽  
Hisayuki Yokoyama ◽  
...  

Abstract The activation of NK cell inhibitory receptors may limit the antitumor efficacy of adoptive autologous and allogeneic NK cell infusions. Recently, we found that exposing malignant cells to the proteosome inhibitor bortezomib upregulated surface expression of death receptors for TRAIL, resulting in significant enhancement of autologous NK cell tumor cytotoxicity in vitro and in vivo. Here we show that NK cells expanded in vitro in the presence of IL-2 and EBV-LCL feeder cells upregulate surface expression of TRAIL which significantly augments bortezomib-induced tumor sensitization to NK cell killing. CD56+/CD3– NK cells were isolated from normal donors by immuno-magnetic bead selection and were co-cultured with irradiated EBV-LCL feeder cells in X-VIVO 20, 10% human AB serum, and 500 IU/ml hrIL-2 for up to 21 days. Depending on culture conditions, a 300 to 10,000 fold increase in NK cell numbers was achieved. Non-expanded and expanded NK cells were analyzed by flow cytometry for the expression of CD56, CD16, TRAIL, FasL, NKG2D, LFA-1, perforin, and granzymes A and B at baseline and ≥ 10 days following in vitro expansion. Chromium release assays were performed to assess fresh vs. expanded NK cell cytotoxicity of renal cell carcinoma (RCC) tumor targets treated with 10 nM bortezomib for 18 hr vs. untreated RCC controls. Freshly-isolated NK cells did not express TRAIL or FasL; in contrast NKG2D, LFA-1, perforin and granzymes A and B were constitutively expressed in fresh NK cells. After expansion, there was a dramatic increase in surface expression of TRAIL and NKG2D; on fresh vs. expanded NK cells from 3 different donors, TRAIL expression increased from 0% to 80.8±15.4% (mean fluorescence intensity [MFI] of TRAIL increased from 6.0±5.1 to 37.9±3.2). The MFI of NKG2D surface expression also increased following NK cell expansion (432.0±70.9 from 48.3±16.3). Expression of LFA-1 and perforin did not change, although there was a small increase in surface and intracellular expression of FasL and granzymes A and B respectively. At a 1:1 effector to target ratio, fresh NK cells lysed 3.4± 2.1% and 5.0± 2.7% of untreated and bortezomib-treated RCC tumor cells respectively. In contrast, there was a dramatic increase in bortezomib-treated tumor susceptibility to killing by expanded NK cells; NK cells expanded for 12-18 days killed 27.6± 9.3% and 55.8± 8.3% of untreated vs. bortezomib-treated RCC tumor cells respectively. Conclusion: In vitro-expanded NK cells are phenotypically and functionally different from non-expanded NK cells. Expanded cells have increased NKG2D and TRAIL expression and greatly enhanced TRAIL-mediated tumor cytotoxicity compared to non-expanded NK cells. Based on these findings, a phase I trial investigating the safety and anti-tumor effects of escalating doses of adoptively-infused ex vivo-expanded autologous NK cells following bortezomib treatment in patients with advanced metastatic tumors and hematological malignancies has recently been initiated. Figure: Freshly isolated and expanded Nk cell lysis of renal cell carcinoma cell line with and without treatment of tumor cells with bortezomib Figure:. Freshly isolated and expanded Nk cell lysis of renal cell carcinoma cell line with and without treatment of tumor cells with bortezomib


Sign in / Sign up

Export Citation Format

Share Document